The relationship between tumor proliferative activity and response to first-line chemotherapy and survival was investigated in 76 advanced breast cancer patients. Proliferative activity was determined by means of Ki-67 immunohistologic staining on primary tumors (55 patients) or at the relapse site (21 patients), and was classified as low ( < or = 25% of stained cells) or high ( > 25% of stained cells). The usual WHO response criteria were used. The median duration of follow-up was 18 months (range 3-58). Forty-seven patients (62%) had tumors with low, and 29 (38%) had tumors with a high rate of proliferative activity. The two groups were well balanced in terms of important variables such as disease-free survival, performance status, age, menopausal status, and the type of first-line chemotherapy (anthracycline-based regimens versus cyclophosphamide-methotrexate-5-fluorouracil). The estrogen receptor (ER) content, measured by means of immunohistochemical assay, was markedly different in the two groups, with 27/47 tumors with low proliferative activity (57%) and 6/29 with high-proliferative activity (21%) being ER positive ( > or = 45% of stained cells) (p = 0.003). Moreover, a significant difference in the metastatic pattern was also evident, with a higher incidence of bone and a lower incidence of soft tissue metastases in the group of patients with tumors with low proliferative activity (p = 0.004). Overall, 10/47 responses (21%: PR = 7, and CR = 3) were observed in the group with a low rate of proliferative activity, versus 14/29 (48%: PR = 9, and CR = 5) in the group with highly proliferative tumors, the difference being statistically significant (p = 0.03). When a multivariate analysis was performed, the only factor that retained independent prognostic significance was the predominant site of disease, particularly soft tissues (p = 0.003). Despite the difference in response rate, when survival analysis was performed according to the Kaplan-Meier method, no significant difference was observed in the two groups, but when the analysis was limited to responsive patients, the median survival observed in those with a low and those with a high rate of proliferation was 35 and 19 months respectively (p = 0.02). The same results were obtained when multivariate survival analysis was carried out using Cox's regression model. These data suggest that there is a link between tumor proliferative activity and response to chemotherapy in advanced breast cancer, and may indicate the need to use more intensive treatments in selected patients with highly proliferative tumors.